Smith P H, Earp J A, DeVellis R
Department of Public Health Education, University of North Carolina at Greensboro 27412-5001, USA.
Womens Health. 1995 Winter;1(4):273-88.
Measuring only the physical markers of violence (e.g., slapping, beating) fails to capture the chronic vulnerability and gendered nature of battered women's experiences. Instruments that measure only observable discrete events may mask the continuous nature of battering and the relation between events and experience. Our approach to measuring battering operationalizes the experiences of battered women rather than the abusive behaviors they encounter. This alternative approach emphasizes the meanings battered women attach to the violence and to battering as an enduring presence in their lives. Focus groups with 22 battered women generated qualitative data for developing scale items (Smith, Tessaro, & Earp, 1995) and a known-groups survey with 185 battered and 204 nonbattered women determined the final scale items. Factor analysis of 40 initial items revealed a strong single-factor solution. The resulting 10-item Women's Experiences with Battering (WEB) Scale demonstrated high internal consistency reliability, was significantly correlated with known-group status, exhibited good construct validity, and was not significantly correlated with a measure of social desirability. The WEB Scale provides researchers with a valid and concise measure for studying relations between battering and health or health behavior, as well as evaluating the impact of interventions on battered women or prevalence.
仅测量暴力的身体指标(如扇耳光、殴打)无法捕捉受虐妇女经历中的长期脆弱性和性别特征。仅测量可观察到的离散事件的工具可能会掩盖殴打行为的持续性以及事件与经历之间的关系。我们测量殴打的方法是将受虐妇女的经历进行量化,而非她们所遭遇的虐待行为。这种替代方法强调了受虐妇女赋予暴力以及殴打行为的意义,因为殴打在她们的生活中是一种持续存在的现象。对22名受虐妇女进行的焦点小组讨论产生了用于制定量表项目的定性数据(史密斯、泰萨罗和厄普,1995年),并对185名受虐妇女和204名未受虐妇女进行了已知群体调查,从而确定了最终的量表项目。对40个初始项目进行的因素分析显示出一个强大的单因素解决方案。由此产生的包含10个项目的“妇女殴打经历量表”(WEB量表)显示出较高的内部一致性信度,与已知群体状态显著相关,具有良好的结构效度,且与社会期望量表无显著相关性。WEB量表为研究人员提供了一种有效且简洁的测量工具,用于研究殴打与健康或健康行为之间的关系,以及评估干预措施对受虐妇女的影响或其患病率。